Stacks.cdc.gov



Supporting Information Figure S1. iEpi phone survey screenshot. Screenshot illustrating the text-based survey on a smartphone during the iEpi Sub-study.Figure S2. eX-FLU log-log plot for nomination social network total degree distribution. The number of nominated individuals by total degree shown on a log-log scale. The network appears scale-free, with the total degree following an approximate power-law distribution. The R2 value for linear trendline fitted to the log-log plot was 0.91.wFigure S3. eX-FLU log-log plot for combined weekly social network total degree distribution. The number of nominated individuals by total degree shown on a log-log scale. The network appears scale-free, with the total degree following an approximate power-law distribution. The R2 value for linear trendline fitted to the log-log plot was 0.85.Table S1. Demographic Characteristics by Enrollment Type for the eX-FLU Studya,bCharacteristicOverallSeedNomineePcNo. of participants590262 (44.4)328 (55.6)Age18.8 (0.04)d18.9 (0.05)d18.6 (0.07)d0.02SexFemale323 (57.9)150 (60.5)173 (55.8)0.27Male235 (42.1)98 (39.5)137 (44.2)RaceWhite364 (67.4)158 (65.6)206 (68.9)0.41Non-white176 (32.6)83 (34.4)93 (31.1)U.S. citizenYes527 (94.6)234 (95.1)293 (94.2)0.64No30 (5.4)12 (4.9)18 (5.8)EmployedYes189 (60.2)84 (60.9)105 (59.7)0.83No125 (39.8)54 (39.1)71 (40.3)Parental education≤ College253 (46.5)114 (47.7)139 (45.6)0.62> College291 (53.5)125 (52.3)166 (54.4)ReligionChristian195 (48.3)86 (47.8)109 (48.6)0.86Non-Christian76 (18.8)36 (20.0)40 (17.9)Non-religious133 (32.9)58 (32.2)75 (33.5)*P≤0.002a Data are N (%) or mean (SE)b See Appendix SA1 for detailed variable definitions.c P-values calculated by Pearson’s Chi-squared or Fisher’s Exact testd Median: 18.0, Interquartile range (IQR): 1.0Table S2. Health Behaviors by Enrollment Type for the eX-FLU Studya,bCharacteristicOverallSeedNomineePcNo. of participants590262 (44.4)328 (55.6)Flu vaccine status, lifetimeEver294 (71.5)127 (72.2)167 (71.1)0.81Never117 (28.5)49 (27.8)68 (28.9)Flu vaccine in 2012-2013Yes162 (45.4)68 (43.9)94 (46.5)0.62No195 (54.6)87 (56.1)108 (53.5)Optimal handwashingYes127 (32.1)56 (30.8)71 (33.2)0.61No269 (67.9)126 (69.2)143 (66.8)Hand sanitizer useEver248 (60.2)110 (59.5)138 (60.8)0.78Never164 (39.8)75 (40.5)89 (39.2)High riskYes100 (24.3)43 (23.8)57 (24.7)0.83No312 (75.7)138 (76.2)174 (75.3)Sleep qualityGood340 (77.5)150 (77.7)190 (77.2)0.90Bad99 (22.6)43 (22.3)56 (22.8)SmokerCurrent11 (2.5)7 (3.6)4 (1.6)0.23Nonsmoker430 (97.5)188 (96.4)242 (98.4)DrinkerYes155 (36.6)62 (33.2)93 (39.4)0.19No268 (63.4)125 (66.8)143 (60.6)ExerciseYes369 (87.2)162 (88.0)207 (86.6)0.66No54 (12.8)22 (12.0)32 (13.4)*P≤0.002a Data are N (%) b See Appendix SA1 for detailed variable definitions.c P-values calculated by Pearson’s Chi-squared or Fisher’s Exact testTable S3. Psychosocial Characteristics by Enrollment Type for the eX-FLU Studya,bCharacteristicOverallSeedNomineePcNo. of participants590262 (44.4)328 (55.6)Perceived stress16.5 (0.3)16.6 (0.4)16.4 (0.4)0.36Shyness33.2 (0.5)33.7 (0.7)32.8 (0.6)0.41GAD severityLow344 (84.1)150 (82.4)194 (85.5)0.40 High65 (15.9)32 (17.6)33 (14.5)LonelinessLow 235 (56.6)112 (60.9)123 (53.3)0.12High180 (43.4)72 (39.1)108 (46.8)Depressiond8.4 (0.2)d8.3 (0.3)e8.5 (0.3)d0.96Interpersonal supportf39.5 (0.2)h39.8 (0.4)i39.3 (0.3)j0.38*P≤0.002a Data are N (%) or mean (SE) b See Appendix SA1 for detailed variable definitions. c P-values calculated by Chi-squared or Fisher’s Exact test for categorical variables or Wilcoxon Rank-Sum test for continuous variables.d P-value calculated using square-root transformed CES-D.d Median: 7.0, IQR: 8.0e Median: 8.0, IQR: 8.0f P-value calculated with power 3 transformed ISEl-12.h Median: 40.0, IQR: 9.0i Median: 41.0, IQR: 9.0j Median: 40.0, IQR: 8.5Table S4. Demographic Characteristics by Intervention Group for the eX-FLU Studya,bCharacteristicOverallControl3-DayNo. of participants590303 (51.4)287 (48.6)Age18.8 (0.04)c18.8 (0.06)c18.7 (0.06)cSexFemale321 (57.7)159 (55.6)164 (60.3)Male235 (42.3)127 (44.4)108 (39.7)RaceWhite364 (67.4)178 (65.2)186 (69.7)Non-white176 (32.6)95 (34.8)81 (30.3)U.S. citizenYes527 (94.6)269 (94.4)258 (94.9)No30 (5.4)16 (5.6)14 (5.2)EmployedYes189 (60.2)99 (64.7)90 (55.9)No125 (39.8)54 (35.3)71 (44.1)Parental education≤ College253 (46.5)140 (51.1)113 (41.9)> College291 (53.5)134 (48.9)157 (58.2)ReligionChristian195 (48.3)101 (50.3)94 (46.3)Non-Christian76 (18.8)37 (18.4)39 (19.2)Non-religious133 (32.9)63 (31.3)70 (34.5)a Data are N (%) or mean (SE)b See Appendix SA1 for detailed variable definitions. c Median:18.0, IQR: 1.0Table S5. Health Behaviors by Intervention Group for the eX-FLU Studya,bCharacteristicOverallControl3-DayNo. of participants590303 (51.4)287 (48.6)Flu vaccine status, lifetimeEver294 (71.5)148 (71.2)146 (71.9)Never117 (28.5)60 (28.9)57 (28.1)Flu vaccine in 2012-2013Yes162 (45.4)80 (44.7)82 (46.1)No195 (54.6)99 (55.3)96 (53.9)Optimal handwashingYes127 (32.1)61 (31.3)66 (32.8)No269 (67.9)134 (68.7)135 (67.2)Hand sanitizer useEver248 (60.2)125 (62.8)123 (57.8)Never164 (39.8)74 (37.2)90 (42.3)High riskYes100 (24.3)50 (24.4)50 (24.2)No312 (75.7)155 (75.6)157 (75.9)Sleep qualityGood340 (77.54)169 (77.9)171 (77.0)Bad99 (22.6)48 (22.1)51 (23.0)SmokerCurrent11 (2.5)4 (1.8)7 (3.2)Nonsmoker430 (97.5)216 (98.2)214 (96.8)DrinkerYes155 (36.6)70 (33.3)85 (39.9)No268 (63.4)140 (66.7)128 (60.1)ExerciseYes369 (87.2)185 (88.1)184 (86.4)No54 (12.8)25 (11.9)29 (13.6)a Data are N (%).b See Appendix SA1 for detailed variable definitions. Table S6. Psychosocial Characteristics by Intervention Group for the eX-FLU Studya,bCharacteristicOverallControl3-DayNo. of participants590303 (51.4)287 (48.6)Perceived stress 16.5 (0.3)16.8 (0.4)16.1 (0.4)Shyness33.2 (0.5)33.7 (0.6)32.7 (0.6)GAD severityLow344 (84.1)165 (82.5)179 (85.7)High65 (15.9)35 (17.5)30 (14.4)LonelinessLow235 (56.6)117 (57.1)118 (56.2)High180 (43.4)88 (42.9)92 (43.8)Depression8.4 (0.2)c8.7 (0.3)d8.2 (0.3)cInterpersonal support39.5 (0.2)e39.2 (0.4)e39.8 (0.3)fa Data are N (%) or mean (SE) b See Appendix SA1 for detailed variable definitions. d Median: 7.0, IQR: 8.0e Median: 7.0, IQR: 7.0h Median: 40.0, IQR: 9.0j Median: 39.0, IQR: 8.0Table S7. Social Network Characteristics by Intervention Group of the eX-FLU Nomination NetworkaCharacteristicOverallControl3-DayNo. of individuals222911771052Indegreeb1.1 (0.02)c1.1 (0.03)c1.1 (0.03)cOutdegreed1.1 (0.06)e1.0 (0.08)e1.1 (0.09)eTotal degreee2.4 (0.06)f2.4 (0.09)f2.5 (0.09)fa Data are mean (SE).b Indegree is the number of nominations an individual received.c Median: 1.0, IQR: 0.0d Outdegree is the number of nomination an individual sent out.e Median: 0.0, IQR: 0.0e Total degree is the number of individuals a person was connected to by nomination or roommate linkages.f Median: 1.0, IQR: 1.0Table S8. Social Network Characteristics by Enrollment Type of the eX-FLU Nomination NetworkaCharacteristicOverallSeedNomineePbNo. of participants590262328Indegreec1.2 (0.05)d0.5 (0.06)e1.7 (0.06)f<0.001*Outdegreeg4.0 (0.2)h3.8 (0.2)i4.3 (0.2)j0.67Total degreek5.2 (0.2)l4.3 (0.2)m6.0 (0.2)h<0.001**P≤0.02a Data are mean (SE).b P-values are calculated by Wilcoxon Rank-Sum tests for continuous variables.c Indegree is the number of nominations an individual received.d Median: 1.0, IQR: 2.0e Median: 0.0, IQR: 1.0f Median: 1.0, IQR: 1.0g Outdegree is the number of nomination an individual sent out.h Median: 4.0, IQR: 5.0i Median: 3.0, IQR: 5.0j Median: 3.0, IQR: 6.0k Total degree is the number of individuals a person was connected to by nomination or roommate linkages.l Median: 4.0, IQR: 6.0m Median: 4.0, IQR: 5.0Table S9. Assortativities for the eX-FLU Studya,bCharacteristicNcAssortativityPdIntervention group22290.35<0.001*Enrollment type5900.070.05Residence hall21720.72<0.001*Residence house21720.46<0.001*iEpi participant5900.22<0.001*Age5900.40<0.001*Sex5580.31<0.001*Race5400.27<0.001*U.S. citizen5570.100.03Parental education5440.110.01Employed3140.150.02Religion4040.080.03Flu vaccine status, lifetime4110.040.44Flu vaccine in 2012-20133570.030.39Optimal handwashing3960.000.99Hand sanitizer use4120.0480.28High risk4120.100.05Sleep quality439-0.090.05Smoker4410.240.02Drinker4230.19<0.001*Exercise423-0.030.64Perceived stress4300.010.76Shyness4110.010.19GAD severity4090.040.36Loneliness4150.010.76Depression405-0.030.60Interpersonal support3830.030.59*P<0.002a See Appendix SA2 for information on assortativity calculations.b See Appendix SA1 for variable definitions.c For each assortativity, the nomination network was reduced to the subset of individuals with relevant data and the links between them.d P-values calculated by a bootstrap method (described in Appendix SA2).Table S10. Centrality Characteristics of the Largest Component of the eX-FLU Studya,bNo. of nominated individuals1827Closeness0.13 (<0.001)Betweenness<0.01 (<0.001)ca Data are mean (SE)b See Appendix SA2 for information on centrality calculations.c Median: 0.0, IQR: 0.004Table S11. Demographic Characteristics of eX-FLU Study Year 2 iEPI participantsa,bCharacteristicOveralliEPINon-iEPIPcNo. of participants590103 (17.5)487 (82.5)Intervention groupControl303 (51.4)49 (47.6)254 (52.2)0.403-Day287 (48.6)54 (52.4)233 (47.8)Enrollment methodSeed262 (44.4)50 (48.5)212 (43.5)0.35Nominee328 (55.6)53 (51.5)275 (56.5)Age18.8 (0.04)d18.7 (0.1)d18.8 (0.05)d0.89SexFemale321 (57.7)68 (66.0)255 (56.0)0.06Male235 (42.3)35 (34.0)200 (44.0)RaceWhite364 (67.4)62 (61.4)302 (68.8)0.15Non-white176 (32.6)39 (38.6)137 (31.2)U.S. citizenYes527 (94.6)99 (96.1)428 (94.3)0.45No30 (5.4)4 (3.9)26 (5.7)EmployedYes189 (60.2)59 (62.8)130 (59.1)0.54No125 (39.8)35 (37.2)90 (40.9)Parental education<College graduate253 (46.5)49 (49.0)204 (46.0)0.58>College graduate291 (53.5)51 (51.0)240 (54.1)ReligionChristian195 (48.3)50 (53.8)145 (46.6)0.47Non-Christian76 (18.8)15 (16.1)61 (19.6)Non-religious133 (32.9)28 (30.1)105 (33.8)a Data are N (%) or mean (SE).b See Appendix SA1 for detailed variable definitions. c P-values calculated by Chi-squared or Fisher’s Exact test for categorical variables or Wilcoxon Rank-Sum for continuous variables.d Median: 18.0, IQR: 1.0Table S12. Health Behaviors of eX-FLU Study Year 2 iEPI participantsaCharacteristicOveralliEpiNon-iEpiPbNo. of participants590103 (17.5)487 (82.5)Flu vaccine status, lifetimeEver294 (71.5)65 (69.9)229 (72.0)0.69Never117 (28.5)28 (30.1)89 (28.0)Flu vaccine in 2012-2013Yes162 (45.4)38 (40.0)124 (47.3)0.22No195 (54.6)57 (60.0)138 (52.7)Optimal handwashingYes127 (32.1)33 (36.3)94 (30.8)0.33No269 (67.9)58 (63.7)211 (69.2)Hand sanitizer useEver164 (39.8)35 (36.5)129 (40.8)0.44Never248 (60.2)61 (63.5)187 (59.2)High riskYes100 (24.3)25 (25.0)75 (24.0)0.85No312 (75.7)75 (75.0)237 (76.0)Sleep qualityGood340 (77.5)82 (80.4)258 (76.6)0.42Bad99 (22.6)20 (19.6)79 (23.4)SmokerCurrent11 (2.5)2 (2.0)9 (2.7)0.69Nonsmoker430 (97.5)100 (98.0)330 (97.4)DrinkerYes155 (36.6)20 (21.1)135 (41.2)<0.001*No268 (63.4)75 (79.0)193 (58.8)ExerciseYes369 (87.2)88 (88.0)281 (87.0)0.79No54 (12.8)12 (12.0)42 (13.0)*P≤0.002 a Data are N (%).b See Appendix SA1 for detailed variable definitions.c P-values calculated by Chi-squared test or Fisher’s Exact test. Table S13. Psychosocial Characteristics of eX-FLU Study Year 2 iEPI participantsa,bCharacteristicOveralliEpiNon-iEpiPcNo. of participants590103 (17.5)487 (82.5)Perceived stress 16.5 (0.3)16.8 (0.6)16.4 (0.3)0.59Shyness33.2 (0.5)33.6 (0.9)33.1 (0.04)0.69GAD severityLow344 (84.1)81 (81.8)263 (84.8)0.47High65 (15.9)18 (18.2)47 (15.2)LonelinessLow235 (56.6)49 (48.5)186 (59.2)0.06High180 (43.4)52 (51.5)128 (40.8)Depressiond8.4 (0.2)e8.3 (0.5)f8.5 (0.2)e0.85Interpersonal supportg39.5 (0.2)h38.8 (0.5)i39.7 (0.3)j0.08*P<0.002a Data are N (%) or mean (SE).b See Appendix SA1 for detailed variable definitions. c P-values calculated by Chi-squared or Fisher’s Exact test for categorical variables or Wilcoxon Rank-Sum for continuous variables. d P-value calculated using square-root transformed CES-D. e Median: 7.0, IQR: 8.0f Median: 7.0, IQR: 7.0g P-value calculated with power 3 transformed ISEL-12.h Median: 40.0, IQR: 9.0i Median: 39.0, IQR: 8.0j Median: 41.0, IQR: 8.0Appendix SA1. Variable derivation and categorization in the eX-FLU study.A number of demographic factors were self-reported by participants during enrollment or on the baseline or exit surveys. Date of birth was used to calculated age in years, which was treated as a continuous measure. Sex, race, and U.S. citizenship were categorized as male or female, White or non-White, and U.S. citizen or non-U.S. citizen. Religious affiliation was collected at baseline and individuals were categorized as identifying as Christian, non-Christian, or non-religious. Parental education was categorized as <= college graduate or > college graduate according to the highest level of educational attainment obtained by a participant’s mother or father. Additionally, participants were classified as employed if they reporting working at all during the Winter 2013 semester on the exit survey. Self-reported information on health behaviors and status were collected on baseline and exit surveys. At baseline, individuals were asked whether they had ever received an influenza vaccine and were dichotomized as ever or never having received a flu vaccine. Individuals who reported ever having received an influenza vaccine were asked about their receipt of the 2012-2013 influenza vaccine at baseline and again at exit, if they did not report receiving the vaccine by baseline. Individuals were categorized as having received or not having received the 2012-2013 influenza vaccine if they ever reported receipt of the vaccine. At baseline, participants also reported on hand hygiene behaviors. Participant’s handwashing behavior was dichotomized as optimal, with handwashing five or more times per day for at least twenty seconds, or not optimal. Individuals who reported an average number of daily hand sanitizer uses greater than zero were categorized as ever using hand sanitizer while participants who responded they never used hand sanitizer or reported zero as their average number of daily hand sanitizer uses were categorized as never using hand sanitizer. Individuals were categorized as being at high risk for a medical complication due to influenza infection if they reported having asthma, reactive airway disorder, type 1 or type 2 diabetes, currently receiving treatment for HIV/AIDS or cancer, previous organ transplant, impaired immunity due to illness or medication, and/or residence or contact with an immunocompromised individual. Participants reporting not having any of the above conditions were classified as not high risk for influenza complications. Sleep quality during the past month was reported as very bad, fairly bad, fairly good, or very good; responses were dichotomized as bad or good. Participants were asked if they currently smoked on a daily basis, and responses were dichotomized as current smoker or nonsmoker. Similarly, participants were categorized as drinkers if they reported drinking alcoholic drinks at least once per week or non-drinkers if they did not report drinking alcoholic drinks at least once per week. Participants were categorized as exercisers if they reported exercising long enough to work up a sweat, get their heart thumping, or get out of breath at least once a week or as non-exercisers if they did not report exercising at least once a week. Psychosocial characteristics were collected at baseline. Participants’ perceived stress level was measured using the Perceived Stress Scale-10 (PSS-10) [1], which measures the degree to which an individual felt out-of-control, emotional, and overwhelmed during the past month, on a 0-4 Likert response scale. Continuous PSS-10 scores were utilized with higher scores indicating higher perceived stress. Shyness was measured using the Revised Cheek and Buss Shyness Scale, containing thirteen items assessed with a 1-5 Likert scale [2]. Responses for all items were summed and treated as a single, continuous measure, with higher scores indicating higher levels of shyness and lower sociability. Generalized Anxiety Disorder (GAD) was assessed using the GAD-7, which consists of seven items assessing the severity of GAD symptoms [3]. Each item is scored from 0-3 based on how frequently an individual reports a feeling (i.e., anxiety, restless, annoyance, etc.) was present during the past two weeks. GAD severity was dichotomized as a score of 0-9 indicating minimal to mild severity or a score of 10-21 indicating moderate to severe severity. Loneliness was measured using the R-UCLA 3-Item Loneliness Scale, which includes three items for which individuals rated their frequency of feelings of loneliness and isolation based on a 1-3 Likert scale [4]. Scores were summed and individuals were dichotomized as having low levels of loneliness (3-5) or high levels of loneliness (6-9). Measures of depression and perceived interpersonal support were additionally assessed. Depressive symptoms were measured with the Center for Epidemiological Studies Depression Scale (CES-D), which measures the frequency of various depressive thoughts and feelings with ten items based on a 1-3 Likert scale [5]. Continuous CES-D scores were utilized, with higher scores indicating higher frequency of depressive symptoms. Perceived interpersonal support levels were measured with the shortened, twelve-item Interpersonal Support Evaluation List (ISEL) [6]. Participants rated their perceived availability of social contacts for various activities and emotional support based on 12 statements using a 1-4 Likert scale. Scores were summed over all items and treated as a continuous measure, with higher values indicating higher perceived social support.References1. Cohen S, Kamarck T, Mermelstein R (1983) A global measure of perceived stress. J Health Soc Behav 24: 385-396.2. Hopko DR, Stowell J, Jones WH, Armento ME, Cheek JM (2005) Psychometric properties of the Revised Cheek and Buss Shyness Scale. J Pers Assess 84: 185-192.3. Spitzer RL, Kroenke K, Williams JB, Lowe B (2006) A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 166: 1092-1097.4. Hughes ME, Waite LJ, Hawkley LC, Cacioppo JT (2004) A Short Scale for Measuring Loneliness in Large Surveys: Results From Two Population-Based Studies. Res Aging 26: 655-672.5. Andresen EM, Malmgren JA, Carter WB, Patrick DL (1994) Screening for depression in well older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale). Am J Prev Med 10: 77-84.6. Cohen SM, R.; Kamarck, T.; Hoberman, H.M. (1985) Measuring the functional components of social support. In: Sarason IGS, B.R., editor. Social Support: Theory, Research and Applications. Dordrecht: Martinus Nijhoff Publishers. pp. 73-94.Appendix SA2. Network characteristics measured in the eX-FLUA network or graph is a collection of nodes and edges (links between nodes), where nodes here represent individuals, and edges represent social network ties (e.g. by nominations, roommate ties, etc.). Below we give definitions for several different network measures used here. Betweenness Centrality. The betweenness centrality [1,2] of a node, i, is a quantitative measure of the extent to which efficient information spread through a network passes through i. It can be defined with respect to the partial betweenness of i for a single pair of other nodes j and k (denoted Pjk(i)), the proportion of all possible shortest paths between j and k that contain i (Equation (1)) [2]. The betweenness centrality of i is equal to the mean of the partial betweennesses of i with respect to all unique pairs of j and k (Equation (2)) [2]. where i, j, k, distinct(1) where i, j, k, distinct(2)Where n is the total number of nodes, represents the number of shortest paths between j and k, and is the number of such paths that pass through i.Closeness Centrality. A node’s closeness centrality refers to the average distance to all other nodes in the network [3,4]. This measure is calculated by taking the reciprocal of a node’s mean geodesic distance, d(i,j), defined as the number of edges in the shortest path between two nodes, between a node i, and all other nodes in the network (j ≠ i) (Equation 3) [3,4]. (3)Assortativity. The assortativity of a undirected network by a given attribute is calculated based on a matrix of mixing probabilities for the attribute of interest (i.e., the probabilities of all combinations of an attribute for two directly linked individuals) [5]. The assortativity coefficient (r) for a given attribute can then be calculated according to:(4)where i represents the categories of an attribute, eii , the diagonal elements of a matrix of mixing probabilities for an attribute, ai, the row sums of the probability matrix, and bi is the column sums of the probability matrix. Bootstrapped Assortativity p-values. Python 2.7 was used to bootstrap assortativity p-values, shown in supplementary Table S9 [6,7]. The ties in the observed network were fixed and then for each attribute, a bootstrapped p-value was calculated by the following process: First, a pool of all observed values for the attribute was created, allowing a given value to appear multiple times. Each node was assigned an attribute by randomly sampling (with replacement) the pool. After all nodes were assigned a random attribute value, assortativity by the attribute was calculated according to Equation (4). This process was performed 10,000 times and the results were compared with the observed assortativity. The number of random-assignment-generated assortativities below observed assortativity was compared to the number above it, and the smaller of the two was used to calculated the assortativity p-value. This smaller count was added to the number of random-assignment-generated assortativities equivalent to the observed assortativity, and this quantity was divided by 10,000 resulting in a proportional value between 0 and 1. This value was then multiplied by 2 to create a two-tailed p-value. The p-value can be interpreted as the probability of obtaining an assortativity as extreme as the observed assortativity under the null hypothesis of zero assortativity.References1. Anthonisse JM (1971) The rush in a directed graph. Stichting Mathematisch Centrum Mathematische Besliskunde: 1-10.2. Freeman LC (1977) A set of measures of centrality based on betweenness. Sociometry: 35-41.3. Beauchamp MA (1965) An improved index of centrality. Behav Sci 10: 161-163.4. Freeman LC (1978) Segregation in Social Networks. Sociological Methods & Research 6: 411-429.5. Newman ME (2003) Mixing patterns in networks. Phys Rev E Stat Nonlin Soft Matter Phys 67: 026126.6. Bickel PJ, Freedman DA (1981) Some Asymptotic Theory for the Bootstrap. Annals of Statistics 9: 1196-1217.7. Efron B (1979) 1977 Rietz Lecture - Bootstrap Methods - Another Look at the Jackknife. Annals of Statistics 7: 1-26. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download